The invention relates to the use of deacetylase genes for producing transgenic plants while employing tissue-specific promoters. In these plants, the development of particular plant parts can be prevented deliberately.
Phosphinothricin (PTC, 2-amino-4-methylphosphinobutyric acid) is an inhibitor of glutamine synthetase (GS). PTC is a xe2x80x9cbuilding blockxe2x80x9d of the antibiotic phosphinothricylalanylalanine. This tripeptide (PTT) is active against Gram-positive and Gram-negative bacteria and also against the fungus Botrytis cinerea. PTT is produced by the Streptomyces viridochromogenes strain Txc3xc494, which is deposited in the Deutsche Sammlung fxc3xcr Mikroorganismen (German collection of microorganisms) under numbers DSM 40736 and DSM 4112 and which is obtainable from this source. It is known from German patent specification 2 717 440 that PTC acts as a total herbicide. The published application (EP-A-0257542) (corresponding to U.S. Pat. No. 5,273,894) describes how a phosphinothricin N-acetyltransferase (pat) gene can be used to produce herbicide-resistant plants. The phosphinothricin N-acetyltransferase which is encoded by the pat gene modifies the PTC which appears intracellularly and detoxifies the herbicide.
The present invention now describes the use of deacetylase genes (dea), whose expression products are able to deacetylate N-acetylphosphinothricin (N-Ac-PTC) and/or N-Ac-PTT intracellularly, and thereby restore their antibiotic activity, for producing female-sterile plants.
An N-acetylphosphinothricin tripeptide deacetylase gene can be isolated from S. viridochromogenes Txc3xc494. The dea gene is located downstream of the pat gene on the already known 4.0 kb BamHI fragment (EP-A-0 257 542) (corresponding to U.S. Pat. No. 5,273,894). This gene is located on a BgIII/BamHI fragment and is fixed precisely by the sequence. The protein sequence is defined by the DNA sequence.
An ATG codon, which is recognized in bacteria and plants, is used as the translation start codon. This gene encodes the last step in the biosynthesis of PTT, i.e. the deacetylation of inactive N-acetylphosphinothricin tripeptide to give the active PTT.
It is known that the specificity of many enzymes is not restricted to one substrate. Thus, the phosphinothricin N-acetyltransferase which is encoded by the pat gene is actually used in PTT biosynthesis for acetylating desmethyl-PTC and, because of its lack of specificity, can be used for detoxifying PTC. By means of overexpressing the dea gene (using suitable promoters or by cloning onto high-copy vectors), an insufficiently specific N-acetyl-PTT deacetylase can now be employed for activating N-acetylphosphinothricin.
Other dea genes can be isolated from E. coli. Thus, it has been found that in E. coli, in contrast to other bacteria (e.g. rhizobias and streptomycetes), no activity can be detected in the so-called pat assay (dissertation of Inge Broer, University of Bielefeld Faculty of Biology, Expression des Phosphinthricin-N-Acetyltransferase-Gens aus Streptomyces viridochromogenes in Nicotiana tabacum (Expression of the Streptomyces viridochromogenes phosphinothricin N-acetyltransferase gene in Nicotiana tabacum), pp. 42-43, 1989) after the pat gene has been cloned into suitable expression vectors (Strauch et al., Gene, 63, 65-74, 1988; Wohlleben et al., Gene, 70, 25-37, 1988). In addition, when present in low copy number in E. coli, the pat gene is unable to confer resistance to PTT since the endogenous deacetylase nullifies the effect of the phosphinothricin N-acetyltransferase. Finally, this deacetylase activity can be demonstrated directly by the efficient inhibition of GS activity which occurs after adding N-acetylphosphinothricin. The deacetylase converts N-Ac-PTC into PTC, which then inhibits the GS in a known manner, as can be measured in a xcex3-glutamyltransferase assay (Bender et al., J. Bacteriol. 129, 1001-1009, 1977). This is due to the possession by E. coli of an endogenous deacetylase activity.
This activity is apparently not present in the argE mutant which is known from the literature (Baumberg, Molec. Gen. Genetics 106, 162-173, 1970). Other E. coli deacetylase mutants are easy to select: following classical (Delixc4x87 et al., Mut. Res. 9, 167-182, 1970; Drake and Baltz, Ann. Rev. Biochem. 45, 11-38, 1976) or Tn5 mutagenesis (Kleckner, Ann. Rev. Genet. 15, 341-404, 1981), such mutants can be recognized on PTT-supplemented minimal medium by the fact that it is only they which are able to grow after having been transformed with a pat gene which is cloned into a low copy number vector.
The E. coli deacetylase gene can therefore be isolated by using conventional methods (Maniatis et al., Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982) to construct a gene library in, for example, the E. coli argE mutant or in a freshly isolated mutant.
Methods for isolating other deacetylase genes can be inferred from that which is described above: e.g. isolating new organisms which are PTT-sensitive despite the presence of a pat gene on a low copy number vector, and subsequently isolating a deacetylase gene.
In a further aspect of the invention, pat and dea genes can be employed together with tissue-specific promoters for deliberately preventing the development of particular plant tissues. An example of a special application is that of producing female-sterile plants.
In plant breeding, the production of hybrid seed depends on avoiding self-fertilization of the parent plant with a high degree of certainty. Male-sterile mutants, which are employed in breeding, occur naturally in many plant species. The molecular mechanism of cytoplasmic male sterility (cms) has not so far been completely clarified. In addition, many crop varieties, such as Beta vulgaris, do not have any cms variant. It is therefore of great interest to agriculture to use the genetic engineering route to generate defined sterile mutants of all the important crop varieties. The company PGS/Belgium has presented such a method in patent application PCT/EP 89/00495. This method is based on destroying the tissue (tapetum) surrounding the pollen parent cells. For this purpose, an RNAse gene is fused to a tapetum-specific promoter (Mariani et al., Nature 347, 737-741, 1990). The exclusive expression of the gene in the tapetum cells ensures that the tissue is destroyed selectively and thereby prevents the mature pollen from being formed. According to this patent, a plant which carries this gene is only able to form seeds following allogamy.
An important disadvantage of this system is the fact that the progeny of this plant are likewise male-sterile and are therefore unable to form any seeds in the field, where they have to rely on self-fertilization. Success in forming seeds is only achieved if the male partner of the cross carries a gene which is able to neutralize the effect of the RNAse in the progeny. According to the abovementioned laid-open patent application, this is supposed to be effected by the barstar gene. In fact, it is only genetically modified, i.e. transgenic partners which can be used in the cross in this context.
Processes for producing female-sterile plants (fs plants), which processes allow transgenic parent plants to be crossed with any partners of the same species, are presented below. This is achieved by the combination of a dea gene under the control of a promoter which is selectively active in the female organs, where appropriate in combination with a constitutively expressed pat gene. The glutamine synthetase in the cells is specifically inhibited, and these cells are caused to die, by applying PTC and/or PTT. An even simpler system comprises producing transgenic plants which only contain one single foreign gene, namely a dea gene under the control of a tissue-specific, in this case female-specific promoter, and then applying N-Ac-PTC and/or N-Ac-PTT to the plant.
To generalize, the invention consequently comprises tissue-specific inhibition with the aid of a deacetylase gene.
1) Plants which are resistant to PTT and/or PTC as a result of Pat activity (e.g. produced as described in EP 0257542 (corresponding to U.S. Pat. No. 5,273,894) or EP 0 242 236) are transformed with a deacetylase gene under the control of the promoter which exhibits tissue-specific activity in plants. Following application of PTT or PTC, expression of the deacetylase gene leads to the activity of phosphinothricin N-acetyltransferase being neutralized in the corresponding tissues. These tissues are then killed selectively whereas the remainder of the plant is resistant.
This system can be simplified by using N-acetylphosphinothricin or N-acetylphosphinothricin-tripeptide. While neither substance is herbicidally active, they are both taken up by plants and transported and not degraded immediately. No deacetylation activity for N-acetylphosphinothricin and N-acetylphosphinothricin-tripeptide has so far been demonstrated in plants. In this way, the above-described 2-gene system can be reduced to a 1-gene system and thereby crucially simplified, as explained in more detail below: any plant can be transformed with a Streptomycetes-derived deacetylase gene under the control of a tissue-specific promoter. Following application of N-acetylphosphinothricin or N-acetylphosphinothricin-tripeptide, the tissue-specific expression leads to the immediate death of the corresponding tissue.
All the described promoters which have been demonstrated to elicit selective expression in particular tissues, preferably the female organs, can be used as tissue-specific promoters. In this connection, the term female organs encompasses the gametophyte and the tissue which surrounds or adjoins it, for example gynoecium (carpels), ovules, placenta, pistil (ovary, style and stigma).
Thus, Robert et al., for example, describe rape-derived stigma-specific promoters (Robert et al., 1994). Pistil-specific promoters have also been described (Sato et al., 1991; Dzelzkalns et al., 1993, WO 94/25613) (corresponding to U.S. Pat. No. 5,859,328).
However, promoters which, while not being specifically active in the female organs, are nevertheless expressed in a tissue which is essential for the development of the functional flower, embryo and seed, are also suitable for use in the process according to the invention.
All newly isolated promoters having similar properties are, of course, also suitable. Apart from tissue-specific promoters, those promoters which are subject to another type of regulation (e.g. temporal, stress-determined or environment-dependent) and which behave in a tissue-specific manner can also be employed.
These processes furthermore make it possible to analyze the differentiation of cell regulation and to produce plants in which the development of particular plant parts has been deliberately prevented. The process can preferably be employed for producing female-sterile plants.